1997 OPEN FORUM Abstracts
IMPACT OF NONINVASIVE VENTILATION ON ICU UTILIZATION
Nazir J. Habib M.D., Kelley B. Jennings RRT, Mark D. Eliasson RRT Kaiser Permanente Medical Center, 975 Sereno Drive. Vallejo, CA 94589-2485
Objectives: Acute respiratory failure is the most common indication for endotracheal intubation and mechanical ventilation which often results in unwanted complications, increased morbidity and mortality rate, and prolonged ICU length of stay. We reviewed the current literature on noninvasive ventilation (NIV) as a first-line intervention in patients with acute respiratory failure (ARF) in correcting gas exchange abnormalities and in avoiding endotracheal intubation (ETI). The approximate success rate from these studies is 70%, with few complications. We describe our experience with NIV and the associated cost savings. Methods: A protocol was developed and the respiratory staff instructed on the use of NIV. Patients admitted to our ICU with ARF were evaluated as potential candidates for NIV and categorized by severity of respiratory failure. Patients who were hemodynamically unstable, exhibited multi-organ dysfunction with sepsis, altered mental status, or hypoxic ARF in whom the clinical condition could not be reversed in 48 hours were not considered candidates. Arterial blood gases, clinical parameters for work of breathing, respiratory rate (RR), and heart rate (HR) were recorded prior to, during and after discontinuation of NIV. The overall success and mortality rate was measured. Results: Over 12 months, we tracked 155 episodes of NIV. 113 episodes of intubation were potentially avoided, for an overall success rate of 73%. There were 32 patients that failed NIV with and overall mortality rate of 47%, compared to no mortality in the successful group. ("Do not intubate" patients were excluded.) The average duration on NIV for the successful group was 1.6 days with an average length of ICU stay of 3 days. The average duration for mechanically ventilated patients with ETI in 1993 for similar case mix was 6.6 days with 7.8 days of ICU stay. The difference between NIV and invasive ventilation duration was 5 days per patient, resulting in a potential reduction 565 ventilator days for the 12 month period. Based on our ICU cost of $1200 per day, $678,000 may have been saved for the year. There were no deaths or complications from not intubating patients and instituting NIV. All deaths following intubation were due to the patient's underlying disease and not to any delay in initiating mechanical ventilation. The average RR in patients successfully supported before and after NIV was 28 and 19 respectively. The average RR before and after in the failed group was 27 and 25 respectively, suggesting that persistent tachypnea after initiation of NIV may be a predictor for failure. HR in the successful group was 110 before NIV and dropped to 98 after NIV. The failed group showed an increase in HR from 108 to 115 suggesting persistent tachycardia may also be a predictor for failure. In 1995 we tracked 196 episodes of NIV with an 85% success rate. In 1996 this success continued. Conclusion: We found NIV successful in most patients with respiratory failure of various etiologies, resulting in a considerable decrease in the number of ventilator days and significant cost savings. We continue to use NIV as a standard of care in ARF.